chienweichang
commited on
Commit
•
72abe5e
1
Parent(s):
cf5c43a
End of training
Browse files- README.md +23 -2
- all_results.json +18 -0
- eval_results.json +13 -0
- generation_config.json +0 -1
- runs/Dec19_03-35-31_8693444b820b/events.out.tfevents.1702961324.8693444b820b.2290.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +270 -0
README.md
CHANGED
@@ -3,9 +3,23 @@ license: apache-2.0
|
|
3 |
base_model: google/mt5-small
|
4 |
tags:
|
5 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
6 |
model-index:
|
7 |
- name: formatted_address
|
8 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
---
|
10 |
|
11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -13,7 +27,14 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
|
14 |
# formatted_address
|
15 |
|
16 |
-
This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
|
18 |
## Model description
|
19 |
|
|
|
3 |
base_model: google/mt5-small
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- cwchang/tw_address_large
|
8 |
+
metrics:
|
9 |
+
- rouge
|
10 |
model-index:
|
11 |
- name: formatted_address
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Summarization
|
15 |
+
type: summarization
|
16 |
+
dataset:
|
17 |
+
name: cwchang/tw_address_large
|
18 |
+
type: cwchang/tw_address_large
|
19 |
+
metrics:
|
20 |
+
- name: Rouge1
|
21 |
+
type: rouge
|
22 |
+
value: 97.0
|
23 |
---
|
24 |
|
25 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
27 |
|
28 |
# formatted_address
|
29 |
|
30 |
+
This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on the cwchang/tw_address_large dataset.
|
31 |
+
It achieves the following results on the evaluation set:
|
32 |
+
- Loss: 0.1388
|
33 |
+
- Rouge1: 97.0
|
34 |
+
- Rouge2: 48.3471
|
35 |
+
- Rougel: 96.996
|
36 |
+
- Rougelsum: 96.9932
|
37 |
+
- Gen Len: 13.7152
|
38 |
|
39 |
## Model description
|
40 |
|
all_results.json
ADDED
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_gen_len": 13.71525,
|
4 |
+
"eval_loss": 0.1387614905834198,
|
5 |
+
"eval_rouge1": 97.0,
|
6 |
+
"eval_rouge2": 48.3471,
|
7 |
+
"eval_rougeL": 96.996,
|
8 |
+
"eval_rougeLsum": 96.9932,
|
9 |
+
"eval_runtime": 210.157,
|
10 |
+
"eval_samples": 4000,
|
11 |
+
"eval_samples_per_second": 19.033,
|
12 |
+
"eval_steps_per_second": 3.174,
|
13 |
+
"train_loss": 0.3973977528091541,
|
14 |
+
"train_runtime": 3972.7609,
|
15 |
+
"train_samples": 40000,
|
16 |
+
"train_samples_per_second": 30.206,
|
17 |
+
"train_steps_per_second": 5.035
|
18 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_gen_len": 13.71525,
|
4 |
+
"eval_loss": 0.1387614905834198,
|
5 |
+
"eval_rouge1": 97.0,
|
6 |
+
"eval_rouge2": 48.3471,
|
7 |
+
"eval_rougeL": 96.996,
|
8 |
+
"eval_rougeLsum": 96.9932,
|
9 |
+
"eval_runtime": 210.157,
|
10 |
+
"eval_samples": 4000,
|
11 |
+
"eval_samples_per_second": 19.033,
|
12 |
+
"eval_steps_per_second": 3.174
|
13 |
+
}
|
generation_config.json
CHANGED
@@ -1,5 +1,4 @@
|
|
1 |
{
|
2 |
-
"_from_model_config": true,
|
3 |
"decoder_start_token_id": 0,
|
4 |
"eos_token_id": 1,
|
5 |
"pad_token_id": 0,
|
|
|
1 |
{
|
|
|
2 |
"decoder_start_token_id": 0,
|
3 |
"eos_token_id": 1,
|
4 |
"pad_token_id": 0,
|
runs/Dec19_03-35-31_8693444b820b/events.out.tfevents.1702961324.8693444b820b.2290.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:559ac459d703b5093c6ad35030a131c389e2df77e67ec4f511fbca0d06e8f04e
|
3 |
+
size 623
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"train_loss": 0.3973977528091541,
|
4 |
+
"train_runtime": 3972.7609,
|
5 |
+
"train_samples": 40000,
|
6 |
+
"train_samples_per_second": 30.206,
|
7 |
+
"train_steps_per_second": 5.035
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,270 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 3.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 20001,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.07,
|
13 |
+
"learning_rate": 4.875006249687516e-05,
|
14 |
+
"loss": 5.8972,
|
15 |
+
"step": 500
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.15,
|
19 |
+
"learning_rate": 4.750012499375032e-05,
|
20 |
+
"loss": 0.9391,
|
21 |
+
"step": 1000
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.22,
|
25 |
+
"learning_rate": 4.625018749062547e-05,
|
26 |
+
"loss": 0.6129,
|
27 |
+
"step": 1500
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.3,
|
31 |
+
"learning_rate": 4.5000249987500626e-05,
|
32 |
+
"loss": 0.4694,
|
33 |
+
"step": 2000
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.37,
|
37 |
+
"learning_rate": 4.3750312484375784e-05,
|
38 |
+
"loss": 0.4011,
|
39 |
+
"step": 2500
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.45,
|
43 |
+
"learning_rate": 4.250037498125094e-05,
|
44 |
+
"loss": 0.369,
|
45 |
+
"step": 3000
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.52,
|
49 |
+
"learning_rate": 4.12504374781261e-05,
|
50 |
+
"loss": 0.334,
|
51 |
+
"step": 3500
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.6,
|
55 |
+
"learning_rate": 4.000049997500126e-05,
|
56 |
+
"loss": 0.3033,
|
57 |
+
"step": 4000
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.67,
|
61 |
+
"learning_rate": 3.875056247187641e-05,
|
62 |
+
"loss": 0.3123,
|
63 |
+
"step": 4500
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.75,
|
67 |
+
"learning_rate": 3.7500624968751565e-05,
|
68 |
+
"loss": 0.2711,
|
69 |
+
"step": 5000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.82,
|
73 |
+
"learning_rate": 3.625068746562672e-05,
|
74 |
+
"loss": 0.264,
|
75 |
+
"step": 5500
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.9,
|
79 |
+
"learning_rate": 3.500074996250188e-05,
|
80 |
+
"loss": 0.255,
|
81 |
+
"step": 6000
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.97,
|
85 |
+
"learning_rate": 3.375081245937703e-05,
|
86 |
+
"loss": 0.2434,
|
87 |
+
"step": 6500
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 1.05,
|
91 |
+
"learning_rate": 3.250087495625219e-05,
|
92 |
+
"loss": 0.2266,
|
93 |
+
"step": 7000
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 1.12,
|
97 |
+
"learning_rate": 3.125093745312735e-05,
|
98 |
+
"loss": 0.2457,
|
99 |
+
"step": 7500
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 1.2,
|
103 |
+
"learning_rate": 3.00009999500025e-05,
|
104 |
+
"loss": 0.2245,
|
105 |
+
"step": 8000
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 1.27,
|
109 |
+
"learning_rate": 2.875106244687766e-05,
|
110 |
+
"loss": 0.223,
|
111 |
+
"step": 8500
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 1.35,
|
115 |
+
"learning_rate": 2.7501124943752816e-05,
|
116 |
+
"loss": 0.2106,
|
117 |
+
"step": 9000
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 1.42,
|
121 |
+
"learning_rate": 2.6251187440627974e-05,
|
122 |
+
"loss": 0.2072,
|
123 |
+
"step": 9500
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 1.5,
|
127 |
+
"learning_rate": 2.5001249937503125e-05,
|
128 |
+
"loss": 0.2031,
|
129 |
+
"step": 10000
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 1.57,
|
133 |
+
"learning_rate": 2.3751312434378282e-05,
|
134 |
+
"loss": 0.2007,
|
135 |
+
"step": 10500
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 1.65,
|
139 |
+
"learning_rate": 2.250137493125344e-05,
|
140 |
+
"loss": 0.1991,
|
141 |
+
"step": 11000
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 1.72,
|
145 |
+
"learning_rate": 2.1251437428128594e-05,
|
146 |
+
"loss": 0.1937,
|
147 |
+
"step": 11500
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 1.8,
|
151 |
+
"learning_rate": 2.000149992500375e-05,
|
152 |
+
"loss": 0.1927,
|
153 |
+
"step": 12000
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 1.87,
|
157 |
+
"learning_rate": 1.8751562421878906e-05,
|
158 |
+
"loss": 0.195,
|
159 |
+
"step": 12500
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 1.95,
|
163 |
+
"learning_rate": 1.7501624918754064e-05,
|
164 |
+
"loss": 0.1935,
|
165 |
+
"step": 13000
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 2.02,
|
169 |
+
"learning_rate": 1.6251687415629218e-05,
|
170 |
+
"loss": 0.1933,
|
171 |
+
"step": 13500
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 2.1,
|
175 |
+
"learning_rate": 1.5001749912504376e-05,
|
176 |
+
"loss": 0.1853,
|
177 |
+
"step": 14000
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 2.17,
|
181 |
+
"learning_rate": 1.375181240937953e-05,
|
182 |
+
"loss": 0.1742,
|
183 |
+
"step": 14500
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 2.25,
|
187 |
+
"learning_rate": 1.2501874906254687e-05,
|
188 |
+
"loss": 0.1842,
|
189 |
+
"step": 15000
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 2.32,
|
193 |
+
"learning_rate": 1.1251937403129845e-05,
|
194 |
+
"loss": 0.1778,
|
195 |
+
"step": 15500
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 2.4,
|
199 |
+
"learning_rate": 1.0001999900005001e-05,
|
200 |
+
"loss": 0.1802,
|
201 |
+
"step": 16000
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 2.47,
|
205 |
+
"learning_rate": 8.752062396880157e-06,
|
206 |
+
"loss": 0.174,
|
207 |
+
"step": 16500
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 2.55,
|
211 |
+
"learning_rate": 7.502124893755312e-06,
|
212 |
+
"loss": 0.1869,
|
213 |
+
"step": 17000
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 2.62,
|
217 |
+
"learning_rate": 6.252187390630469e-06,
|
218 |
+
"loss": 0.1786,
|
219 |
+
"step": 17500
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 2.7,
|
223 |
+
"learning_rate": 5.002249887505625e-06,
|
224 |
+
"loss": 0.1794,
|
225 |
+
"step": 18000
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 2.77,
|
229 |
+
"learning_rate": 3.7523123843807808e-06,
|
230 |
+
"loss": 0.18,
|
231 |
+
"step": 18500
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 2.85,
|
235 |
+
"learning_rate": 2.502374881255937e-06,
|
236 |
+
"loss": 0.1791,
|
237 |
+
"step": 19000
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 2.92,
|
241 |
+
"learning_rate": 1.2524373781310935e-06,
|
242 |
+
"loss": 0.1684,
|
243 |
+
"step": 19500
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 3.0,
|
247 |
+
"learning_rate": 2.4998750062496878e-09,
|
248 |
+
"loss": 0.168,
|
249 |
+
"step": 20000
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 3.0,
|
253 |
+
"step": 20001,
|
254 |
+
"total_flos": 2890815009177600.0,
|
255 |
+
"train_loss": 0.3973977528091541,
|
256 |
+
"train_runtime": 3972.7609,
|
257 |
+
"train_samples_per_second": 30.206,
|
258 |
+
"train_steps_per_second": 5.035
|
259 |
+
}
|
260 |
+
],
|
261 |
+
"logging_steps": 500,
|
262 |
+
"max_steps": 20001,
|
263 |
+
"num_input_tokens_seen": 0,
|
264 |
+
"num_train_epochs": 3,
|
265 |
+
"save_steps": 5000,
|
266 |
+
"total_flos": 2890815009177600.0,
|
267 |
+
"train_batch_size": 6,
|
268 |
+
"trial_name": null,
|
269 |
+
"trial_params": null
|
270 |
+
}
|